Moderated mediation - get details (intercepts) for simple slopes plot

[Lukas Wallrich]

Dear all,

I am running the model below to test for moderated serial mediation. With that, I can get the total effects at different levels of the moderator, which I would take to be the slopes for a simple slopes plot of the total effect. However, could you advise how I can get the intercepts for these plots? I.e. in this model the intercepts for UniSelfDiscB at different levels of SixtFNegContBlack.c?

Any advice would be much appreciated.

All the best,

Lukas

--

Path.model <- "

             #Model

             UniSelfDiscB ~ c*SixthFPosContBlack.c  + dd*ActiveAvoidanceBlack_1 +cx*SixthFPosContBlack.c:SixtFNegContBlack.c+cn*SixtFNegContBlack.c #+ bbb*ContSelfEfficBlack_1

             ContSelfEfficBlack_1 ~ a*SixthFPosContBlack.c +ax*SixthFPosContBlack.c:SixtFNegContBlack.c + an*SixtFNegContBlack.c

             ActiveAvoidanceBlack_1 ~ b*ContSelfEfficBlack_1 #+ aa*SixthFPosContBlack.c

              #Covars

             UniSelfDiscB ~ srcINT

             

             #Mean of moderator

             SixtFNegContBlack.c ~ SixtFNegContBlack.c.mean*1

             

             #Variance of moderator

             SixtFNegContBlack.c ~~ SixtFNegContBlack.c.var*SixtFNegContBlack.c

             

             #Indirect

             MMM := a*b*dd

             #M := a*bbb

             #indirect := MMM+M

             

             #Indirect effects conditional on moderator (a + ax*ModValue)*chain

             indirect.SDbelow := (a + ax*(SixtFNegContBlack.c.mean-sqrt(SixtFNegContBlack.c.var)))*b*dd # + (a + ax*(SixtFNegContBlack.c.mean-sqrt(SixtFNegContBlack.c.var)))*bbb

             indirect.mean := (a + ax*(SixtFNegContBlack.c.mean))*b*dd # + (a + ax*(SixtFNegContBlack.c.mean)))*bbb

             indirect.SDabove := (a + ax*(SixtFNegContBlack.c.mean+sqrt(SixtFNegContBlack.c.var)))*b*dd #+ (a + ax*(SixtFNegContBlack.c.mean-sqrt(SixtFNegContBlack.c.var)))*bbb

            #Direct effects conditional on moderator (c + cx*ModValue)

            direct.SDbelow := c + cx*(SixtFNegContBlack.c.mean-sqrt(SixtFNegContBlack.c.var)) 

            direct.mean := c + cx*(SixtFNegContBlack.c.mean)

            direct.SDabove := c + cx*(SixtFNegContBlack.c.mean+sqrt(SixtFNegContBlack.c.var))

            

             #Total effects

             total.SDbelow := direct.SDbelow + indirect.SDbelow

             total.mean := direct.mean + indirect.mean

             total.SDabove := direct.SDabove + indirect.SDabove

             

                          

             #Proportion mediated conditional on moderator  

             prop.mediated.SDbelow := indirect.SDbelow / total.SDbelow

             prop.mediated.mean := indirect.mean / total.mean

             prop.mediated.SDabove := indirect.SDabove / total.SDabove

  "

[Terrence Jorgensen]

how I can get the intercepts for these plots? I.e. in this model the intercepts for UniSelfDiscB at different levels of SixtFNegContBlack.c?

You would need to add intercepts to the plot, then define simple intercepts just as you define simple slopes (assuming that syntax works).  Find formulas here:

http://quantpsy.org/interact/interactions.htm

Terrence D. Jorgensen

Assistant Professor, Methods and Statistics

Research Institute for Child Development and Education, the University of Amsterdam

http://www.uva.nl/profile/t.d.jorgensen